. The Canadian field-naturalist. 156 The Canadian Field-Naturalist Vol. 97 This map was constructed from 238 individual series lag values. Figure 3 is an isoplethic map of a surrogate used to help interpret relative amplitudes of the cycle in differ- ent areas. To derive this map, the proportions of the "no change" category noted at each grid cell location each year have been totalled for the seventeen year period. This value is an imperfect but useful surrogate for amplitude because it can be assumed that observers were likely to note "no change" more often for areas whose
. The Canadian field-naturalist. 156 The Canadian Field-Naturalist Vol. 97 This map was constructed from 238 individual series lag values. Figure 3 is an isoplethic map of a surrogate used to help interpret relative amplitudes of the cycle in differ- ent areas. To derive this map, the proportions of the "no change" category noted at each grid cell location each year have been totalled for the seventeen year period. This value is an imperfect but useful surrogate for amplitude because it can be assumed that observers were likely to note "no change" more often for areas whose populations fluctuated less obviously. All 290 values were used to generate this map. Figure4 is an isoplethic map of a surrogate statistic representing degree of non-random trend in the resid- uals of the detrended series. In most of the map, this can be considered to correspond to mean degree of cycle strength; in a few areas, however(notably the St. Lawrence Valley), trend was high despite lack of sig- nificant years periodicity. This situation arises because the surrogate used here is the Durbin-Watson statistic, which measures trend only (and not cyclicity perse). Apparently, some parts of the study area were either undergoing long term population irruptions (unlikely) or were simply cycling in periods greatly varying from the year national mean. Areas of the map that tested "no trend" at the level of signifi- cance are labelled such; of the remaining areas, lower values signify clearer trend. All 290 values were used to generate this map. Despite the many possible sources of error in deriv- ing these summary maps, it is felt they do a reasonably good job of representing the original data. A further validity check on the data of Figure 2 was carried out by averaging the values mapped on a province-by- province basis. The results are listed in Table 1. These results correlate very highly (r = ) with Bulmer's. FiGi re2. Geographic variation in S
Size: 1806px × 1383px
Photo credit: © Book Worm / Alamy / Afripics
License: Licensed
Model Released: No
Keywords: ., bookauthorottawafieldnaturalistsclub, bookcentury1900, bookcolle
